Conceptos Básicos
Machine learning applied to blockchain data is a growing and relevant topic of interest, with a focus on anomaly detection, cryptocurrency price prediction, and address classification.
Resumen
The content provides a systematic mapping study on machine learning applied to blockchain data. It covers the objectives, methodology, results, and conclusions of the study. The study identifies key use cases such as anomaly detection, cryptocurrency price prediction, and address classification. It also explores different blockchains analyzed, data sources used, dataset sizes, availability of data, and types of machine learning models applied.
Abstract:
- Blockchain technology has garnered significant attention in both literature and practice.
- Machine learning applied to blockchain data is a relevant and growing field.
- The study aims to systematically review the state of the art in this area.
Introduction:
- Blockchain technology offers transparency with all transactions recorded publicly.
- Machine learning can analyze blockchain data for patterns and predictions.
- The study focuses on identifying research gaps in machine learning on blockchain data.
Methodology:
- Conducted a systematic mapping study following established guidelines.
- Research questions focused on topics related to machine learning on blockchain.
- Used various database sources for primary research.
Results:
- Majority of papers focused on anomaly detection use case (49.7%).
- Bitcoin was the most analyzed blockchain (47.1%).
- Multiple data sources were used for analysis (29.6%).
- Dataset sizes varied with over 1 million data points being common (31.4%).
Use Cases:
Address Classification:
- Focuses on de-anonymization or actor identification.
- Algorithms used include NB, AdaBoost, SVM, LR, RF.
- Datasets consisted of over 1 million data points covering different periods.
Estadísticas
A dataset consisting of more than 1.000.000 data points was used by 31.4% of the papers.
Citas
"The results confirm that ML applied to blockchain data is a relevant and a growing topic" - Study Conclusion